Question: htseq-count result for strand specific RNA-seq data using three "-s" mode
gravatar for guoql.karen
3.8 years ago by
guoql.karen10 wrote:

Hey there!

Could anyone give some explainations for the htseq-count result here:

  1. DATA

  2. SAM file produced by mapping RNA-seq data (fr-secondstrand) to the genome.

  3. GTF filE


  • use samtools to sort the BAM file by name, and converse it to SAM file

  • use htseq-count and sorted BAM file to count reads for each of the genes in GTF. Meanwhile, I want to see the three "-s" options results using my strand specific RNA-seq data.

Command like this: "htseq-count -i gene_id -s no accepted_hits_sorted_name.sam $gtf > both_strand.txt"

The other two change the "-s no" to "-s yes" and "-s reverse".

RESULTS (I merged the result as below:)

ID s=no s=yes s=reverse

EPlOSAG00000000069 4 0 4

EPlOSAG00000000035 10 10 0

EPlOSAG00000000022 0 18 0

EPlOSAG00000000053 58 69 1

EPlOSAG00000000065 239 2 243

EPlOSAG00000000072 39 80 0


The "-s no" means count the reads number regarless of the strand. Besides, "yes" and "reverse" means count the strand specific reads number according to the GTF annotation. Then the third row and fourth row should be added together equal to the second row (I can understand the number is not simply equal here because of the complexed conting mode. But it will not exceed the "no" options too much I think). However many of the results are out of this rules. For "EPlOSAG00000000072", the "yes" option counts much more reads comparing to "no" option. who can tell me the reasons?

Thank you very much!

rna-seq next-gen • 1.7k views
ADD COMMENTlink modified 3.8 years ago by Carlo Yague4.9k • written 3.8 years ago by guoql.karen10
gravatar for Carlo Yague
3.8 years ago by
Carlo Yague4.9k
Carlo Yague4.9k wrote:

If some reads overlap two features on opposed strands, they will be discarded as ambiguous when "-s no" but they will be kept if you count them 'strand-specifically'.

For more information, see the overlap resolution mode of HTseq-count here

ADD COMMENTlink modified 3.8 years ago • written 3.8 years ago by Carlo Yague4.9k

OK! Thank you for your comments! It's helpful!

ADD REPLYlink written 3.8 years ago by guoql.karen10
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1112 users visited in the last hour